Skip to content

Exceptionally high carrier mobility in hexagonal diamond

FieldValue
Journalnpj Computational Materials
Published2026-09-19
AuthorsZirui He, Shang‐Peng Gao, Meng Chen
DOI10.1038/s41524-026-02330-4

Ab initio transport calculations predict an electron mobility of 28,473 cm² V⁻¹ s⁻¹ in hexagonal diamond at 300 K along the c axis. The model gives 12,339 cm² V⁻¹ s⁻¹ perpendicular to c, while hole mobilities reach 6,024 and 6,000 cm² V⁻¹ s⁻¹ along and perpendicular to c. These are phonon limited predictions, not measured device mobilities.

At 300 K, the same framework gives cubic diamond mobilities of 2,540 cm² V⁻¹ s⁻¹ for holes and 2,152 cm² V⁻¹ s⁻¹ for electrons. The hexagonal electron mobility is 5.7 times higher perpendicular to c and 13.2 times higher along c.

CarrierDirection relative to cMobility at 300 K
HolePerpendicular6,000 cm² V⁻¹ s⁻¹
HoleParallel6,024 cm² V⁻¹ s⁻¹
ElectronPerpendicular12,339 cm² V⁻¹ s⁻¹
ElectronParallel28,473 cm² V⁻¹ s⁻¹

The predicted values exceed the paper’s cited reference points of 1,820 cm² V⁻¹ s⁻¹ for holes in germanium and 8,900 cm² V⁻¹ s⁻¹ for electrons in GaAs. The hole result agrees with an earlier theoretical prediction for hexagonal diamond, while the electron result is substantially higher than that earlier value.

A GGA calculation gave hexagonal diamond hole and electron mobilities of 2,151 and 1,680 cm² V⁻¹ s⁻¹. These values agree with prior theoretical results of 2,290 and 1,664 cm² V⁻¹ s⁻¹, respectively.

The calculations compare crystalline hexagonal diamond with cubic diamond. Hexagonal diamond has AB stacking along its c axis, whereas cubic diamond has ABC stacking.

QUANTUM ESPRESSO handled geometry optimization and density functional perturbation theory phonon calculations. The work used the PBEsol functional, a 100 Ry plane wave cutoff, and fully relativistic norm conserving pseudopotentials.

Quasi particle band structures were calculated with the one shot G0W0 method in YAMBO. The exchange self energy and dielectric function used 60 Ry and 22 Ry cutoffs, respectively. WANNIER90 constructed maximally localized Wannier functions for the band structures and densities of states.

EPW interpolated electronic Hamiltonians, dynamical matrices, and electron phonon matrix elements onto ultrafine k and q grids. The transport window extended 0.3 eV from each band edge. Carrier mobilities were solved with the Boltzmann transport equation, while phonon mode contributions were analyzed with self energy relaxation time calculations.

An alternative interpolation scheme and an ABINIT calculation confirmed the predicted transport trend. The study therefore reports calculations rather than measurements on synthesized devices.

Band structure does not explain the mobility difference

Section titled “Band structure does not explain the mobility difference”

The mobility effective masses at 300 K are comparable between the two diamond phases.

Carrierh diamond, perpendicular to ch diamond, parallel to cc diamond, average
Hole0.4770.6780.489
Electron0.4620.2610.371

The values are given in electron rest mass units. The densities of states near the band edges are also comparable.

Interband and intervalley channels do not account for the large mobility difference. Excluding the hexagonal split off hole band from the calculation raises hole mobility by only about 2%. Intervalley scattering contributes approximately 10% to electron mobility reduction in both phases at room temperature.

The impurity calculation used randomly distributed, singly charged Coulomb centres at a concentration of 10¹⁷ cm⁻³. Without phonon scattering at 300 K, hexagonal diamond gives hole mobilities of 1,904 and 1,521 cm² V⁻¹ s⁻¹ across and along c. Its electron values are 2,005 and 2,421 cm² V⁻¹ s⁻¹. Cubic diamond gives 2,121 cm² V⁻¹ s⁻¹ for holes and 1,994 cm² V⁻¹ s⁻¹ for electrons. Similar impurity results support the conclusion that weaker electron phonon scattering causes the mobility difference.

Acoustic deformation potential scattering below 30 meV dominates around room temperature. For holes, the total relaxation time is 1.83 ps in hexagonal diamond and 0.70 ps in cubic diamond, a ratio of 2.62. The transverse acoustic limited value is 6.59 ps versus 1.19 ps, giving a larger ratio of 5.55. Optical phonon limited times are nearly equal at 5.50 and 5.38 ps.

The suppression of transverse acoustic scattering follows from hexagonal symmetry. The TA-a mode is forbidden for in-plane transitions by the kz = 0 mirror plane. Both transverse acoustic modes are forbidden for scattering along the c axis by Δ5 symmetry. The TA-b mode remains allowed for in-plane scattering and contributes more strongly than TA-a.

The valence band maximum has in-plane px py character. Out of plane longitudinal acoustic vibrations and interlayer shear or breathing modes therefore have weak spatial overlap with the hole wavefunction. This mismatch reduces their deformation potentials.

Only optical modes 7, 8, and 10 produce non-negligible hole deformation potentials in hexagonal diamond. Their total optical scattering contribution remains comparable to cubic diamond because the other optical branches have weak coupling or are forbidden by symmetry.

Electron transport and wavefunction decoupling

Section titled “Electron transport and wavefunction decoupling”

Transverse acoustic scattering is almost absent for electrons in hexagonal diamond. Its electron relaxation time is 72.7 ps, compared with 0.66 ps in cubic diamond. The ratio is 110. Total electron relaxation times are 3.49 and 0.44 ps, a ratio of 7.94. Longitudinal acoustic limited times are 4.92 and 1.89 ps, while optical limited times are 17.2 and 6.21 ps.

The conduction band minimum lies at K and has D3h symmetry. TA-a scattering is forbidden within the kz = 0 plane. TA-b scattering is also forbidden within the same ΓKHA plane because the electronic states and phonon have incompatible odd parity. Vertical mirror planes intersect the c axis at 60° intervals, further restricting the available scattering regions.

The conduction band state consists of antibonding px py orbitals concentrated within cages formed by carbon atoms and bonds. Phonon scattering potentials lie mainly near atoms and bonds, so their overlap with the conduction state is small. This real space decoupling lowers the electron phonon matrix elements for both transverse and longitudinal acoustic modes.

Optical mode 10 strongly distorts the cages and has a large deformation potential. Its effect remains limited near room temperature because optical intervalley scattering dominates optical intravalley scattering in both phases.

Temperature dependence and thermal stability

Section titled “Temperature dependence and thermal stability”

Mobility does not follow one power law between 150 and 700 K. The increasing occupation of optical phonons changes the dominant scattering mechanism. Their energies lie between 138 and 160 meV, and their contribution rises with temperature.

The electron mobility advantage decreases with temperature. At 700 K, the hexagonal to cubic ratio is 3.8 perpendicular to c and 7.6 parallel to c. Hole mobility also changes anisotropy as optical scattering becomes more important.

The self energy relaxation time approximation assumes complete loss of carrier momentum after each scattering event. It therefore underestimates hole mobility along c by 16% at 300 K and 3.5% at 700 K.

The G0W0 calculation gives hexagonal diamond a wide bandgap of 4.55 eV. A molecular dynamics simulation used an NPT ensemble at 1000 K and 1 atm. It ran for 10,000 steps, or 10 ps, with a 1,280 atom supercell and indicated thermal and structural stability.

Prior predictions cited by the paper give a room temperature thermal conductivity near 2,000 W m⁻¹ K⁻¹. By comparison, boron arsenide has an ambipolar mobility near 1,600 cm² V⁻¹ s⁻¹ and a cited thermal conductivity near 1,300 W m⁻¹ K⁻¹.

The calculations identify hexagonal diamond as a candidate for high frequency and high power electronics. Electron transport benefits most from alignment with the c axis, while hole transport becomes more transverse at higher temperatures. Device design must therefore establish both the crystal phase and the transport orientation.

The paper reports no measured carrier mobility for hexagonal diamond. Its transport results describe phonon limited behavior, while the impurity model uses random Coulomb centres without detailed defect structures. It does not resolve how specific defects alter the band structure or phonon dispersion.

The 10 ps molecular dynamics run supports stability at 1000 K but does not establish long duration stability. The next experimental test should measure mobility along and across c in phase confirmed hexagonal diamond. Until those measurements exist, 28,473 cm² V⁻¹ s⁻¹ remains a computational target rather than an available material specification.

View Original Abstract

Hexagonal diamond (h-diamond), or Lonsdaleite, is a promising wide-bandgap semiconductor known for its high thermal conductivity and hardness. Based on ab initio calculations, we demonstrate its exceptionally high carrier mobilities. At room temperature, the hole mobilities along the ⊥ c and ∥ c directions are 6000 and 6024 cm 2 V −1 s −1 , respectively, while the corresponding electron mobilities reach 12,339 and 28,473 cm 2 V −1 s −1 . These values are significantly superior to those of most known semiconductors, including cubic diamond. The small effective masses in h-diamond are comparable to those in the cubic phase, which cannot explain its substantially higher mobilities. Instead, two underlying mechanisms are uncovered. First, selection rules enforced by the symmetry of h-diamond significantly suppress scattering, particularly for transverse acoustic phonons, which predominate in the cubic phase around room temperature. Secondly, the spatial mismatch between the electronic wavefunctions and phonon-induced scattering potentials leads to real-space electron-phonon decoupling, which manifests as the suppression of out-of-plane polarised longitudinal acoustic scattering for holes, and a systematic weakening of acoustic scattering for electrons.